A brain-tumor model utilizing stereotactic implantation of a permanent cannula.

A tumor model involving stereotactically implanted culture-reared tumor cells is presented. Stainless steel cannulas were stereotactically and permanently implanted into the caudate nucleus of 30 rats. The animals were separated into two groups. In Group I, 15 animals received a 10-microliters injection containing 10(6) C6 glioblastoma cells (five rats), 10(6) Walker 256 breast carcinoma cells (five rats), or cell medium (five rats). The coordinates were A(+1.5), L(+3.0), and DV(-5.0). In Group II, the coordinates were changed to A(+1.0), L(+3.0), and DV(-5.0) and the same number of rats received a 1-microliter injection containing 10(5) cells of each tumor in an attempt to produce more focal tumors. Two weeks after implantation, brain sections were stained with cresyl violet and a subset was stained for glial fibrillary acid protein (GFAP). A computerized morphometric analysis system was used to quantify tumor size. In Group I, the mean C6 tumor areas (+/- standard error of the mean) at specific coordinates were (in sq mm): A(+4.7) 0.4 +/- 0.2; A(+3.7) 3.5 +/- 1.1; A(+2.7) 5.7 +/- 1.7; A(+1.7) 9.5 +/- 2.3; A(+0.7) 7.5 +/- 3.2; A(-0.3) 3.7 +/- 2.9; and A(-1.3) 0.3 +/- 0.3. A nearly identical tumor mass and extension into the brain was produced in rats injected with Walker 256 cells. Similar C6 tumor areas were indicated in adjacent sections stained with cresyl violet and GFAP. Tumor was found in the caudate nucleus in all 10 rats, but not in the nucleus accumbens, fornix, or hippocampus. In Group II animals, tumor magnitude and extension into the brain were greatly reduced. The 10(6) cells in the 10-microliters volume was the most reliable tumor load for obtaining uniform tumors in different animals. The similarity of tumor distribution across different animals was indicated by the low variance of tumor area at specific anteroposterior coordinates. Reproducible and well-circumscribed caudate nucleus tumors were produced using this stereotactic procedure.

Attempts to increase intratumoral blood flow in the rat solid Walker 256 tumor by the use of the perfluorocarbon emulsion fluosol-DA.

We have examined the effect of the perfluorocarbon emulsion, Fluosol-DA 20% (FDA), on blood flow in rats bearing an advanced solid Walker 256 tumor implanted s.c. Blood-FDA exchange in unanesthetized rats maintained under 100% oxygen was accomplished by simultaneous arterial withdrawal and i.v. infusion until the hematocrit was less than 4%. Control rats were maintained under 100% oxygen but did not undergo any exchange. Regional blood flow studies in tumors of control and FDA-exchanged rats were performed using [14C]iodoantipyrine and quantitative autoradiography. FDA-blood exchange did not increase flow to the whole tumor. Similarly, the pattern of regional flow within the tumor, which was determined in histologically distinct areas--including dense and normocellular, necrotic and peripheral zones invading into muscle and connective tissue--was not substantially altered. Flow to cerebral tissue was increased two-fold, although flow to normal tissues including temporalis muscle, skin and diaphragm was not altered. These results show that FDA-blood exchange does not enhance vascular flow in solid Walker 256 tumor implanted s.c. in the rat.

Increases in brain tumor and cerebral blood flow by blood-perfluorochemical emulsion (Fluosol-DA) exchange.

It has been suggested that oxygen-carrying blood substitutes, perfluorochemical (PFC) emulsions, can increase blood flow and oxygen delivery to poorly perfused tumor regions. Local cerebral blood flow was measured in male Wistar rats bearing intracranial Walker 256 tumor with and without blood-PFC exchange using [14C]iodoantipyrine (IAP) and quantitative autoradiographic techniques. The exchange transfusion was performed in two groups of awake animals breathing 100% oxygen: (a) complete blood-PFC exchange, hematocrit 4%; and (b) partial blood-PFC exchange, hematocrit 20-25%. The tissue/blood partition coefficient for IAP was determined in a separate set of experiments under identical conditions and was used in calculating blood flow. Cerebral blood flow increased approximately 2-fold following complete blood-PFC exchange and 1.5-fold by the partial exchange. A similar 1.5-fold increase in flow was measured in intraparenchymal tumors following partial exchange; however, a flow increase was not identified in the meningeal extension of the tumors. The increase in cerebral blood flow is consistent with an autoregulatory response of the central nervous system vasculature to maintain an adequate supply of oxygen to central nervous system tissue. Presumably, the increase in blood flow to the intracerebral tumor reflects the autoregulatory response of the host tissue. The effect of blood-PFC exchange on blood flow and drug delivery to tumor may depend on the particular tumor and its site of growth (host tissue). The tissue/blood partition coefficient for IAP increased from 0.8 to 1.0 and 1.4 following partial and complete blood-PFC exchange, respectively. This change in the partition coefficient reflects the change in the intravascular fraction of IAP that is bound to plasma proteins. The enhanced therapeutic effect that has been reported in some experimental tumor models may result from a higher tissue/blood equilibrium distribution ratio (due to reduced plasma protein binding) resulting in a higher tissue exposure to certain drugs following PFC administration.

Dissociation of 5-fluorouracil-induced DNA fragmentation from either its incorporation into DNA or its cytotoxicity in murine T-lymphoma (S-49) cells.

We have shown previously (W. B. Parker and P. Klubes, Cancer Res., 45:4249-4256, 1985) that uridine (10 microM) enhanced the cytotoxicity of 5-fluorouracil (FUra) in cultured mouse T-lymphoma (S-49) cells. Here we show, by the use of colony formation assays, that approximately 50% of the cytotoxicity of FUra plus uridine could be prevented by the simultaneous administration of thymidine (2.5 to 10 microM). In order to explain our observation of a thymidine-preventable component of the cytotoxicity of the FUra plus uridine combination, we examined the incorporation of FUra into DNA. The DNA from FUra-treated S-49 cells was purified by cesium chloride gradient centrifugation and degraded to nucleosides by DNase I and Crotalus atrox snake venom. 5-[3H]-Fluoro-2'-deoxyuridine was not detected by high-pressure liquid chromatography in the hydrolysate of DNA from S-49 cells treated with 1.0 microM [3H]FUra, 1.0 microM [3H]FUra plus 10 microM uridine, or 2.4 microM [3H]FUra. In contrast, 5-[3H]fluoro-2'-deoxyuridine was detected in the DNA of L1210 cells treated with cytotoxic concentrations of either [3H]FUra or 5-[3H]fluoro-2'-deoxyuridine. Thus incorporation of FUra into the DNA of S-49 cells treated with cytotoxic concentrations of FUra was shown to be minimal or insignificant. Using alkaline elution techniques, however, fragmentation of the DNA was detected in S-49 cells treated with 1.0 microM FUra, 1.0 microM FUra plus 10 microM uridine, or 2.4 microM FUra (115-, 107-, and 159-rad equivalent single strand breaks, respectively). Most of the DNA fragmentation caused by FUra could be prevented by the inclusion of 2.5 microM thymidine with FUra during the incubation. Similar amounts of DNA fragmentation occurred with 1.0 microM FUra in either the presence or absence of 10 microM uridine. Because 1.0 microM FUra plus 10 microM uridine was more cytotoxic than 1.0 microM FUra alone, these results indicated that the enhancement of FUra cytotoxicity by uridine was not related to increased fragmentation of DNA.

Comparison of the bioavailability of uridine in mice after either oral or parenteral administration.

We compared the bioavailability of uridine (Urd) (350 and 3500 mg/kg) administered either as a single SC injection or by gavage, in male CD8F1 mice. Plasma samples were analyzed for Urd and uracil (Ura) using high-pressure liquid chromatography. After Urd (3500 mg/kg, SC), plasma Urd levels peaked at 4900 microM and then declined to pretreatment levels (less than 10 microM) within 6 h. Plasma Ura concentrations peaked at 1400 microM and then declined initially more slowly than Urd. After Urd (3500 mg/kg, PO) plasma levels of Urd were fairly constant (range 33-82 microM) for up to 8 h and had returned to pretreatment levels at 16 h. Plasma Ura concentrations paralleled Urd, but were approximately ten-fold higher. Areas under the concentration-time curve for Urd showed that the bioavailability of Urd after PO administration was 7% of that after SC administration. After Urd (350 mg/kg, SC) Urd levels peaked at 210 microM returning to pretreatment levels within 2 h. Plasma Ura levels reached a peak with 300 microM and then declined initially more slowly than those of Urd. After Urd (350 mg/kg, PO) plasma Urd levels were not perturbed, although Ura levels peaked at 50 microM after which they declined and could no longer be detected at 4 h. These data indicate that the bioavailability of Urd (350 or 3500 mg/kg) was lower when given PO than when it was administered by SC injection; and Urd (3500 mg/kg) PO resulted in prolonged and relatively constant plasma Urd levels compared with Urd (3500 mg/kg) SC. These results suggest that Urd PO should be compared with parenterally administered Urd in attempts to increase the therapeutic index of 5-fluorouracil and of antimetabolite inhibitors of de novo pyrimidine biosynthesis.

Enhancement by uridine of the anabolism of 5-fluorouracil in mouse T-lymphoma (S-49) cells.

Uridine enhances the growth inhibition due to 5-fluorouracil (FUra) in a cultured mouse T-cell lymphoma (S-49). Using colony formation assays we found that cytotoxicity produced by 24-h continuous exposure to FUra (0.5 to 3.5 microM) was increased more than two-fold by simultaneous exposure to 10 microM uridine. Studies were undertaken to explain the mechanism by which uridine enhanced the cytotoxicity of FUra in S-49 cells. Uridine (10 microM) increased by about 50% both the anabolism of 1.0 microM [3H]FUra to acid-soluble metabolites and the incorporation of 1.0 microM [3H]-FUra into RNA. However, the incorporation of 1.0 microM [3H]FUra into these fractions was less than that seen with 2.4 microM [3H]-FUra, a dose which was equitoxic to 1.0 microM [3H]FUra plus 10 microM uridine. High-pressure liquid chromatography analysis of the acid-soluble metabolites of FUra did not show any selective change in specific FUra nucleotides, which could explain the increased cytotoxicity associated with 10 microM uridine. In addition, 5-fluoro-2'-deoxyuridine monophosphate levels and the amount of [3H]FUra which was incorporated into the alkali-stable, acid-insoluble fraction were not increased by uridine. Uridine (10 microM) inhibited de novo pyrimidine biosynthesis by 70%, while 5-phosphoribosyl-1-pyrophosphate levels were unchanged. Presumably, the inhibition of de novo pyrimidine biosynthesis decreased orotic acid levels and allowed more FUra to be anabolized to 5-fluorouridine monophosphate via orotate phosphoribosyl transferase. Furthermore, 2.4 microM FUra inhibited the incorporation of [3H]deoxyguanosine into DNA by 50% after 24 h of incubation. In contrast, 1.0 microM FUra plus 10 microM uridine did not inhibit the incorporation of [3H]deoxyguanosine into DNA. The data suggested that there was a qualitative difference in the mechanism by which 1.0 microM FUra plus 10 microM uridine killed S-49 cells as compared to 2.4 microM FUra alone, and that the enhancement by uridine of the cytotoxicity of FUra was due, in part, to the increased anabolism of FUra to ribonucleotides.

Use of uridine rescue to enhance the antitumor selectivity of 5-fluorouracil.

We examined the ability of uridine to increase the therapeutic index of 5-fluorouracil (FUra) against C57BL/6 X DBA/2 F1 mice bearing a Day 1 B16 melanoma or L1210 leukemia. FUra (400, 600, or 800 mg/kg, i.p.) followed in 24 hr by a 5-day s.c. infusion with uridine (5 g/kg/day, s.c.) was compared with the maximum tolerated dose of FUra (200 mg/kg, i.p.) plus a 5-day infusion with 0.9% NaCl solution. High-dose FUra plus delayed infusion with uridine was more effective than FUra (200 mg/kg) in inhibiting the growth of the B16 melanoma. High-dose FUra plus uridine rescue was, however, no more effective than FUra (200 mg/kg) in increasing the survival times of mice bearing the L1210 leukemia. To see if uridine rescue from FUra toxicity correlated with effects against a sensitive normal tissue, bone marrow nucleated cellularity of normal, non-tumor-bearing mice was monitored after drug treatment. In mice treated with FUra (200 mg/kg) followed in 24 hr by a 5-day infusion with either uridine (5 g/kg/day) or 0.9% NaCl solution, there was not as great a decrease in cellularity at the nadir with uridine, and, in addition, uridine accelerated recovery as compared to 0.9% NaCl solution. Furthermore, uridine (5 g/kg/day), but not thymidine (dThd) (5 g/kg/day) or 2'-deoxyuridine (dUrd) (5 g/kg/day), had a sparing effect on the depression in bone marrow nucleated cellularity seen at the nadir on Day 4 after Fura (200 mg/kg). The specificity of uridine to rescue mice from the lethal toxicity of the related fluorinated pyrimidines, 5-fluorouridine and 5-fluoro-2'-deoxyuridine, was also examined. Mice were treated with 5-fluorouridine (250 mg/kg, i.p.) followed in 24 hr by a 5-day infusion with uridine (1, 5, or 10 g/kg/day), dThd (1, 5, or 10 g/kg/day), or dUrd (1 or 5 g/kg/day). Uridine (1, 5, or 10 g/kg/day) rescued mice from the lethal toxicity of 5-fluorouridine, whereas dThd or dUrd was ineffective. Similarly, a 5-day infusion with uridine, but not dThd or dUrd, rescued mice from the lethal toxicity of 5-fluoro-2'-deoxyuridine (1800 mg/kg, i.p.).

Uridine rescue from the lethal toxicity of 5-fluorouracil in mice.

To determine the relationship between 5-fluorouracil (FUra) toxicity and its RNA- and DNA-directed actions we examined the ability of continuous SC infusions with uridine (Urd), thymidine (dThd), or deoxyuridine (dUrd) to rescue mice from the lethal toxicity of FUra. Male B6D2F1 mice were treated with a single IP injection of FUra (800 mg/kg) followed in 24 h by a 5-day infusion with either 0.9% NaCl or Urd (0.1, 1, 5, or 10 g/kg/day). Survivors were then followed up for 30 days after FUra treatment. Urd (1, 5, or 10 g/kg/day) rescued mice from the lethal toxicity of FUra, whereas Urd (0.1 g/kg/day) was as ineffective as 0.9% NaCl as a rescue agent. With variable doses of FUra followed in 24 h by a Urd infusion (5 g/kg/day) for 5 days. Urd rescued mice treated with FUra (400, 600, or 800 mg/kg) but was ineffective against higher doses of FUra (1,000 or 1,200 mg/kg). Mice treated with FUra (800 mg/kg) followed in 24 h by a 5-day infusion with either dThd (1, 5, or 10 g/kg/day) or a dUrd (1 or 5 g/kg/day) could not be rescued from the lethal toxicity of FUra. In all experiments deaths occurred between 6 and 12 days after FUra. These results, which demonstrate a specificity for Urd, but not for either dThd or dUrd, for rescuing mice from the lethal toxicity of FUra, suggest the importance of the RNA- rather than the DNA-directed actions of FUra as a determinant of its toxicity in mice.

Effect of concurrent calcium leucovorin infusion on 5-fluorouracil cytotoxicity against murine L1210 leukemia.

We examined the effect of concurrent SC infusion of calcium leucovorin (LV) on the action of 5-fluorouracil (FUra) against mouse L1210 leukemia implanted either SC or IP. Mice bearing the SC tumor treated with FUra (100 mg/kg, IP, day 1) plus infusion with either LV (11.5 mg . kg-1 . day-1, days 1-4), or 0.9% NaCl (days 1-4) resulted in an identical increase in median lifespan (ILS) of 28%. Similar experiments with FUra (100 mg/kg) plus LV infusion (115 mg . kg-1 . day-1) or FUra (200 mg/kg) plus LV infusion (115 mg . kg-1 . day-1) resulted in 50% and 59% ILS, respectively, which were not different from that obtained with the same doses of FUra plus 0.9% NaCl infusion. Mice bearing the IP tumor treated with FUra (100 mg/kg, IP, day 1) plus infusion with either LV (11.5 mg . kg-1 . day-1, days 1-4) or 0.9% NaCl (days 1-4) had an identical 56% ILS. Similar experiments with FUra (100 mg/kg) plus LV infusion (115 mg . kg-1 . day-1) or FUra (200 mg/kg) plus LV infusion (115 mg . kg-1 . day-1) resulted in 67% and 94% ILS, respectively, which were not different from those obtained with the same doses of FUra plus 0.9% NaCl infusion. Treatment of normal mice with FUra (200 mg/kg, IP, day 0) plus LV infusion (115 mg . kg-1 . day-1, days 0-3) was no more toxic than FUra plus 0.9% NaCl infusion, judging by similar transient decreases in body weight and no mortality. The data indicate that concurrent infusion with the LV failed to enhance the action of FUra against the mouse L1210 leukemia.

Biochemical and pharmacologic study of therapeutic synergism with cyclophosphamide plus methotrexate in murine L1210 leukemia.

The combination of cyclosphosphamide (CP) plus methotrexate (MTX) was found to be therapeutically synergistic against the L1210 ascites tumor. After inoculation with 1 x 10(6) L1210 cells to (C57BL/6 x DBA/2)F1 mice, ip administration of CP (200 mg/kg on Day 5) plus MTX (15 mg/kg on Days 5, 7, 9, and 11) resulted in a significant increase in mean lifespan, compared to optimal treatment with either CP or MTX alone. The contribution of the single, simultaneous dose of CP plus MTX on Day 5 to therapeutic synergism was examined. The combination of CP plus MTX on Day 5 produced a greater decrease in tumor cell numbers than either drug alone. Measurements of the time course of inhibition and recovery of 3H-deoxyuridine incorporation into DNA showed that within 48 hours, recovery of DNA synthesis in small intestine and bone marrow was almost complete after either CP, MTX, or CP plus MTX. In contrast, tumor, which had recovered within 48 hours after MTX, still remained greater than 90% inhibited after eight CP or CP plus MTX. Measurements of bone marrow nucleated cellularity after therapy showed that the single, simultaneous dose of CP plus MTX was no more toxic to bone marrow than CP alone, although it was more toxic than MTX alone. No alterations in the distribution of either MTX or alkylating metabolites of CP could be detected in tumor or normal tissues when CP and MTX were administered simultaneously.

Studies of the effects of treatment with 5-flourouracil alone and in combination with either cyclophosphamide, methotrexate, or methyl-CCNU on hepatic drug-metabolizing enzymes in murine L1210 leukemia.

Studies have been conducted on liver microsomal enzymes of B6D2F(1) mice bearing a Day 4 L1210 ascites tumor after treatment with a single intraperitoneal injection of 5-fluorouracil (FUra) alone, and in combination with either cyclophosphamide (CP), methotrexate (MTX), or methyl-CCNU (MeCCNU). FUra (200 mg/kg) alone did not alter either ethylomorphine N-demethylase, aniline hydroxylase, or neotetrazolium reductase activities as compared to untreated nontumor-bearing mice on Days 2, 4 and 6 after treatment. FUra (50 mg/kg)d plus CP (200 mg/kg) decreased neotetrazolium reductase activity 17% on Days 4 and 6 after treatment and the enzyme activity then returned to control levels on Day 8. Ethyl morphine N-demethylase and aniline hydroxylase, or neotetrazolium reductase activities on Days 1, 2, 4 and 6 after treatment. FUra (50 mg/kg) plus MeCCNU (21 mg/kg) did not affect either ethylmorphine N-demethylase or aniline hydroxylase activities on Days 1, 2, 4, 6 and 8 after treatment. However, neotetrazolium reductase activity was decreased 15% on Day 2 and then returned to control levels on Day 4. These results indicate that treatment of mice bearing the L1210 ascites tumor with an optimal single does of FUdra alone and in combination with either CP, MTX or MeCCNU does not have marked or prolonged effects on liver microsomal drug-metabolizing enzyme activities.

Effects of 5-fluorouracil on human colon carcinoma and solid rat Walker 256 carcinosarcoma: evaluation as in vitro predictors of clinical response.

Several biochemical effects of 5-fluorouracil (5-FU) including inhibition of the incorporation of 3H-deoxyuridine (3H-UdR) into DNA, inhibition of ribosome formation, and formation of 5-fluoro-2'-deoxyuridine-5'-monophosphate (FdUMP) were examined in samples of human colon carcinomas to determine if any of these drug effects might have predictive value as a reliable guide to 5-FU therapy. For comparison, the solid rat Walker 256 carcinosarcoma, which is only minimally responsive to 5-FU, was also studied. For each of the biochemical effects of 5-FU measured in the various samples of Walker 256 tumors, the responses were consistent and varied within a narrow range. In contrast, the formation of FdUMP from 5-FU and the degree of inhibition of the incorporation of 3H-UdR into DNA by 5-FU were extremely variable in the population of human colon carcinomas examined. In all human tumors examined, 5-FU caused a reduction in the formation of ribosomes, but even in the absence of 5-FU, the total amount of ribosome synthesis was so low that it makes measurement difficult to quantitate. Based on our data, a study might be warranted to determine if there is a correlation between FdUMP formation and responsiveness of colon carcinoma to 5-FU therapy.

Distribution of anticancer agents in spontaneous animal tumors. II. Distribution of gallium in canine lymphosarcoma.

The physiologic disposition of pharmacologic doses of gallium was studied in control dogs and dogs with spontaneous lymphosarcoma. Gallium 67 (67Ga) was administered iv with carrier gallium added at a dose of 8 mg/kg. About 50% of the injected dose was excreted in the urine by 48 hours (mostly in the first 12 hr), whereas negligible amounts were excreted in bile. The distribution of 67Ga in normal tissues was similar in control and tumor-bearing dogs. The tissue-to-plasma concentrations of gallium were considerably greater than 1 in the kidney cortex, bone marrow, bone, small intestine, and liver 6-72 hours after administration of the drug. At comparable time periods, tissue-to-plasma ratios of gallium were less than 1 in skeletal muscle and brain. In dogs with lymphosarcoma there was neither selective uptake nor selective retention of gallium in comparison to most normal tissues. In fact, several normal tissues, particularly kidney cortex and bone marrow, concentrated gallium greatly in excess of tumors. Qualitatively similar findings were obtained in a dog with malignant melanoma. These findings were contrary to what one would predict from reports showing that carrier-free 67Ga is selectively concentrated in various human and animal tumors. This indicates the need for more extensive studies of the physiologic disposition of pharmacologic (antitumor) doses of gallium in humans and appropriate animal models.